Owning the architecture: auth and identity flows, multi-tenancy, how data moves and where it's allowed to live, performance, and what happens in production
Running what you build. You'll be on-call for your own work from early on, before there's a dedicated platform to lean on, which means owning the infrastructure, CI, deployment, and observability behind it
Working on top of external platforms you can't change, and being straight about which risks are yours to manage and which aren't
Partnering closely with the internal teams whose services you depend on: asking clearly, agreeing on interfaces, and finding ways to keep moving instead of waiting around to be unblocked